Aligning conservation goals: are patterns of species richness

Animal Biodiversity and Conservation 24.1 (2001)
91
Aligning conservation goals: are patterns
of species richness and endemism
concordant at regional scales?
T. H. Ricketts
Ricketts, T. H., 2001. Aligning conservation goals: are patterns of species richness and endemism concordant
at regional scales? Animal Biodiversity and Conservation, 24.1: 91–99.
Abstract
Aligning conservation goals: are patterns of species richness and endemism concordant at regional scales?—
Biodiversity conservation strategies commonly target areas of high species richness and/or high endemism.
However, the correlation between richness and endemism at scales relevant to conservation is unclear; these
two common goals of conservation plans may therefore be in conflict. Here the spatial concordance between
richness and endemism is tested using five taxa in North America: butterflies, birds, mammals, amphibians, and
reptiles. This concordance is also tested using overall indices of richness and endemism (incorporating all five
taxa). For all taxa except birds, richness and endemism were significantly correlated, with amphibians, reptiles,
and the overall indices showing the highest correlations (rs = 0.527–0.676). However, “priority sets” of
ecoregions (i.e., the top 10% of ecoregions) based on richness generally overlapped poorly with those based
on endemism (< 50% overlap for all but reptiles). These results offer only limited support for the idea that
richness and endemism are correlated at broad scales and indicate that land managers will need to balance
these dual, and often conflicting, goals of biodiversity conservation.
Key words: Conservation goals, Species richness, Endemism, Regional scales.
Resumen
Ajuste de los objetivos de conservación ¿Son concordantes a escala regional los patrones de riqueza de especies
y de endemismos?— Las estrategias de conservación de la biodiversidad se centran habitualmente en áreas con
una gran riqueza de especies y/o un alto nivel de endemicidad. Sin embargo, la correlación entre riqueza y
endemismo a escala relevante para la conservación es poco clara; por consiguiente, estos dos objetivos
comunes de los planes de conservación pueden entrar en conflicto. En este trabajo se estudia la concordancia
espacial entre riqueza y endemismo en Norteamérica utilizando cinco taxones: mariposas, aves, mamíferos,
anfibios y reptiles. Esta concordancia se estudia también empleando índices globales de riqueza y endemismo
(incorporando los cinco taxones). Para todos los taxones, excepto para las aves, riqueza y endemismo aparecen
correlacionados significativamente, mostrando para los anfibios y reptiles una alta correlación de todos los
índices (rs = 0.527–0.676). Sin embargo, las “actuaciones prioritarias” de las ecoregiones (por ejemplo, el 10%
de ecoregiones de vanguardia) basadas en la riqueza de especies normalmente se solapan poco con las basadas
en endemismos (< 50% de solapamiento para todos los taxones excepto para los reptiles). Estos resultados
apoyan limitadamente la idea de que riqueza y endemismo están correlacionados a gran escala e indica que los
gestores del territorio deberán tener en cuenta estos objetivos duales, y a menudo en conflicto entre sí, de
conservación de la biodiversidad.
Palabras clave: Objetivos de conservación, Riqueza de especies, Endemismo, Escala regional.
(Recieved: 4 X 01; Final acceptance: 10 X 01)
Taylor H. Ricketts, Center for Conservation Biology, Dept. of Biological Sciences, Gilbert Hall / 371 Serra Mall,
Stanford Univ., Stanford, CA 94305–5020 (USA).
ISSN: 1578–665X
© 2001 Museu de Zoologia
Ricketts
92
Introduction
It is well recognized by conservation biologists
that there are limited resources available to
address intensifying anthropogenic threats to
biodiversity (EHRLICH & WILSON, 1991; MYERS et al.,
2000). Geographic priorities must therefore be
established, so that these resources and effort
can be allocated to areas with high biodiversity
value, such as high species richness and/or
endemism (C EBALLOS et al., 1998; O LSON &
DINERSTEIN, 1998). While in theory this is a sound
strategy, its implementation has encountered
two major difficulties. First, a lack of high–quality
species distribution data, especially at broad
scales, has made it difficult to identify priority
areas with confidence (WILLIAMS & GASTON, 1994).
Second, there is frequently a difference of opinion
among conservationists over which aspects of
biodiversity are most important in setting
priorities. Some authors have emphasized species
richness, while others argue that areas of high
endemism should be targeted most (PRENDERGAST
et al., 1993; KERR, 1997; CEBALLOS et al., 1998).
A popular response to this first problem has
been to propose indicator taxa: well–studied
groups of organisms whose richness patterns can
be used as surrogates for other taxa or for
overall species richness. Many recent studies have
either proposed indicator taxa (e.g., PEARSON &
CASSOLA, 1992), assumed them to indicate overall
richness and based conservation plans on them
(e.g., SCOTT et al., 1993), or tested their utility
directly (e.g., DAILY & EHRLICH, 1996; CARROLL &
PEARSON, 1998; RICKETTS et al., 1999a; RICKETTS et
al., in press). To date, tests of indicator taxa for
species richness have produced mixed results,
suggesting the utility of this conservation tool
depends on context, taxon, and scale (WEAVER,
1995).
Even if suitable indicator taxa can be found,
however, the second problem remains. Priorities
set on the basis of species richness may not
successfully conserve areas of high endemism,
which are clearly important to biodiversity
conservation at any scale. Data on endemism are
typically less available than on species richness,
and patterns of endemism are thus less well
understood (BIBBY, 1992; KERR, 1997). Therefore,
biologists still have a relatively poor understanding
of whether patterns of species richness and
endemism are concordant, and thus whether
these two common goals of conservation plans
are in conflict or alignment.
This second problem is addressed here, using
a large North American dataset to examine the
concordance of richness and endemism patterns
in five animal taxa (butterflies, birds, mammals,
reptiles, and amphibians). Two specific questions
are asked. First, are levels of species richness and
endemism correlated across the United States
and Canada? this correlation is tested for each
taxon individually as well as for indices of overall
richness and endemism that incorporate all five
taxa. Second, to what extent do areas selected
for conservation priority on the basis of richness
overlap with areas selected on the basis of
endemism? Answers to these questions will help
determine whether the two primary goals of
biodiversity conservation plans will tend to
reinforce or compete with each other for limited
resources.
Methods
Species data
The species distribution data are based on the
110 ecoregions of the continental United States
and Canada (fig. 1). These ecoregions were first
developed by Ricketts et al (RICKETTS et al., 1999b),
and are based largely on three established
ecoregion mapping projects (ESWG 1995; GALLANT
et al., 1995; O MERNIK , 1995). Ecoregions are
relatively coarse biogeographic divisions of a
landscape that delineate areas with broadly
similar environmental conditions and natural
communities. They are nested within eight major
biomes in North America (fig. 1). Because of the
complexity with which environmental and
ecological factors vary across a landscape,
ecoregion boundaries are necessarily approximate
and represent areas of transition rather than
sharp divisions.
RICKETTS et al. (1999b) compiled presence/
absence data for butterflies, birds, mammals,
reptiles, and amphibians on these ecoregions.
The same dataset was used after performing
further checks for quality and accuracy. From
presence/absence data, the number of species
(hereafter “richness”) and the number of endemic
species (hereafter “endemism”) were calculated
of each taxon in every ecoregion. Following
Fig. 1. A. Map of the 110 terrestrial ecoregions of the United States and Canada; B. Map showing
the eight biomes represented by these ecoregions.
Fig. 1. A. Mapa de las 110 ecorregiones terrestres de Estados Unidos y Canadá; B. Mapa que
muestra ocho biomas representados por estas ecorregiones.
93
Animal Biodiversity and Conservation 24.1 (2001)
A
B
Temperate broadleaf and mixed forests
Mediterranean scrub and savanna
Temperate coniferous forests
Xeric shrublands/deserts
Temperate grasslands/savannas
Boreal forest/taiga
Flooded grasslands
Tundra
Ricketts
94
Table 1. Spearman rank correlations between richness and endemism for the five animal taxa
considered. Results given for all 110 ecoregions, and for each biome separately: * Significance
level at p < 0.05 (missing entries indicate that in the corresponding taxon, no endemic species
are found in any ecoregion of the corresponding biome); All. Includes six ecoregions from
minor biomes that are not included in any of the biome analyses.
Tabla 1. Correlaciones del rango de Spearman entre riqueza y endemismo para los cinco taxones
considerados. Los resultados se indican para cada una de las 110 ecorregiones y para cada bioma
por separado: * Nivel de significancia para p < 0,05 (los datos que faltan indica que en el
correspondiente taxón no se han encontrado especies endémicas en ninguna ecorregión del
bioma que le corresponde); All. Incluye seis ecorregiones de biomas pequeños que no están
incluidas en ninguno de los biomas analizados.
Biome
Butterflies
Birds
0.304*
0.011
All
Temperate broadleaf
–0.153
Temperate coniferous
0.298*
Overall
indices
n
0.527*
0.676*
0.588*
110
0.153
–0.112
0.795*
0.430
0.635*
17
0.471*
–
0.186
0.632*
0.526*
0.479*
30
–
0.407
–0.098
0.704*
0.796*
0.504*
16
0.327
0.412
0.218
–0.127
0.788*
0.753*
8
Temperate grasslands
Xeric shrublands
Mammals Amphibians Reptiles
Boreal forest / Taiga
–
0.584*
0.333
–
–
0.442
17
Tundra
–
0.398
–0.212
–
–
0.047
16
RICKETTS et al. (1999b), a species to be endemic in
an ecoregion was counted if it either (i) was
found in no other ecoregion, including Mexico
and other continents or (ii) occupied a range
totaling less than 50,000 km2 (BIBBY, 1992). Thus
species with exceptionally small ranges that
crossed an ecoregion boundary were considered
endemics in both ecoregions.
To examine more general patterns of
biodiversity, overall indices of richness and
endemism that incorporate information from all
five taxa were also calculated. The richness index
was defined as
5
1/5
5
i
Ri / Ti
where Ri is the richness of taxon i in the
ecoregion, and Ti is total number of species of
taxon in the database (SISK et al., 1994; RICKETTS
et al., 1999a). This index normalizes the richness
of each taxon by the number of North American
species in that taxon and then averages those
fractions across all five taxa. It therefore weights
taxa evenly, preventing speciose groups from
dominating measures of overall richness.
The endemism index is defined as
5
1/5
5E / R
i
i
i
where Ei is the number of endemic species of
taxon i in the ecoregion, and Ri is as above. This
index computes, for each taxon, the fraction of
species in an ecoregion that is endemic there,
and then averages these fractions across all five
taxa. Again, the index thus normalizes counts of
endemics by the taxon’s richness in each
ecoregion.
Analyses
Correlation between richness and endemism
measures were tested using Spearman rank
correlations, because data were seldom normally
distributed (ZAR, 1999). Since ecoregions vary
widely in area (fig. 1, RICKETTS et al., 1999b) and
both richness and endemism are typically
expected to increase with area (ROSENZWEIG, 1995),
any correlations found may be driven by these
area effects. To examine this possibility, the
Spearman rank correlation between the richness
and endemism measures and ecoregion area was
computed. Finally, to examine whether the
degree of concordance between richness and
endemism differs among biomes, the richness/
endemism correlations were tested for each
biome independently (fig. 1, table 1).
To determine the overlap between richness–
based and endemism–based priority sets of
ecoregions, the ecoregions in the top decile were
identified (i.e., 90th percentile and above) for
each measure. The percent overlap of these sets
for each taxon, and for the overall indices were
then calculated (PRENDERGAST et al., 1993). The
top decile of 110 ecoregions typically contains
95
Animal Biodiversity and Conservation 24.1 (2001)
A
B
3
8
7
Endemism
6
2
5
4
3
1
2
1
0
0
0
C
50 100 150 200 250 300
D
7
Endemism
6
0
50 100 150 200 250 300
0
10 20 30
25
20
5
15
4
3
10
2
5
1
0
0
0
20
40
60
80
100 120
E
40 50 60 70
F
18
0.1
16
Endemism
14
0.08
12
0.06
10
8
0.04
6
4
0.02
2
0
0
0
20
40 60
Richness
80 100 120
0
0.1
0.2
0.3
Richness
0.4
Fig. 2. Relationship between richness and endemism across all 110 ecoregions. Each circle represents
an ecoregion: A. Butterflies; B. Birds; C. Mammals; D. Amphibians; E. Reptiles; F. Overall richness and
endemism indices. Dashed lines in panel f delineate the top decile on each axis; note only two
ecoregions lying above both lines (i.e., in the upper right quadrant). These are the only two ecoregions
that are members of both richness-based and endemism–based priority sets, and they are coded in
green in figure 3F.
Fig. 2. Relación entre riqueza y endemicidad en las 110 ecorregiones. Cada círculo representa una
ecorregión: A. Mariposas; B. Aves; C. Mamíferos; D. Anfibios; E. Reptiles; F. Índices globales de riqueza
y endemicidad. Las líneas discontinuas en la figura F delimitan el decilo superior de cada eje; obsérvese
que únicamente dos ecorregiones se encuentran por encima de ambas líneas (por ejemplo en el
cuadrante superior derecho). Éstas son las dos únicas ecorregiones que optan a la vez por las actuaciones
prioritarias basadas en la riqueza y en el endemismo, están indicadas en negro en la figura 3F.
Ricketts
96
11 ecoregions. In some cases, however, ties
between the 11th–ranked ecoregion and those
ranked below it forced inclusion of more than
11 in the priority set. Overlap between richness
and endemism in these cases was calculated by
dividing the number of shared ecoregions by the
number of ecoregions in the smaller of the two
priority sets (PRENDERGAST et al., 1993).
Results
Across all North American ecoregions, species
richness and endemism were in general positively
correlated (table 1, top row). For all taxa except
birds, richness and endemism were significantly
correlated, with amphibians, reptiles, and the
overall indices showing the highest correlations.
There is a large amount of scatter in bivariate
plots for all taxa, however (fig. 2), indicating a
low degree of predictive power in these
relationships.
Richness and endemism for most taxa were
not significantly correlated with ecoregion area
(table 2). The only three significant relationships
found (i.e., involving endemism in butterflies,
Table 2. Spearman rank correlations
between ecoregion area and measures of
richness and endemism for the five taxa
and for the overall indices: * Significance
level p < 0.05 (n = 110).
Tabla 2. Correlaciones de rango de
Spearman entre área de ecorregión y
medidas de la riqueza y endemicidad para
los cinco taxones y para la totalidad de
índices: * Nivel de significación p < 0,05
(n = 110).
Taxon
Richness
Endemism
Butterflies
0.09
-0.21*
Birds
0.07
0.19*
Mammals
0.15
-0.26*
Amphibians
Reptiles
Overall indices
0.05
-0.08
-0.02
0.00
0.09
-0.16
Table 3. Percent overlap between priority sets of ecoregions based on richness and endemism.
Tabla 3. Porcentaje de solapamiento entre prioridades de ecorregiones basado en riqueza y endemicidad.
Amphibians
Reptiles
Overall
indices
11
11
11
12
13
15
13
15
11
3 (27%)
3 (27%)
5 (45%)
7 (64%)
2 (18%)
Butterflies
Birds
Richness set
11
11
Endemism set
18
4 (36%)
Overlap
Mammals
Fig. 3. Maps showing the distribution of, and overlap between, richness and endemism priority
sets. Light gray ecoregions are in the top decile for richness, medium gray ecoregions are in the
top decile for endemism, and black ecoregions are in the top decile for both: A. Butterflies; B.
Birds; C. Mammals; D. Amphibians; E. Reptiles; F. Overall richness and endemism indices.
Fig. 3. Mapas que muestran la distribución y la coincidencia de las acciones prioritarias en
riqueza y endemicidad: Gris claro, ecorregiones situadas en el decilo superior en cuanto a
riqueza; Gris medio, ecorregiones situadas en el decilo superior en cuanto a endemicidad; Negro,
ecorregiones situadas en el decilo superior para ambas prioridades: A. Mariposas; B. Aves; C.
Mamíferos; D. Anfibios; E. Reptiles; F. Índices globales de riqueza y endemicidad.
97
Animal Biodiversity and Conservation 24.1 (2001)
A
B
C
D
E
F
98
birds, and mammals) were weak and inconsistent
in their sign (table 2). Therefore, the correlation
results in table 1 are unlikely to be caused by the
commonly–expected effects of area on richness
and endemism.
When correlations were tested within each
biome independently, the results generally
reflected those found using all ecoregions
(table 1). Amphibians, reptiles, and the overall
indices again tended to show strong correlations
in all but the tundra and taiga biomes.
Correlations for butterflies, birds and mammals,
which showed weak or no correlation using all
ecoregions, remained generally non–significant
in the biome–by–biome analyses.
Overlap between richness-based and endemism–
based priority sets were generally low, varying
between 27% (birds and mammals) to 64%
(reptiles) (table 3). In addition, priority ecoregions
for richness and endemism were often found on
opposite sides of the continent and often in
different biomes (fig. 3).
Discussion
These results offer mixed support for the idea
that richness and endemism patterns are
correlated at broad scales. On one hand, two taxa
and the overall indices showed quite strong and
consistent correlations across the 110 ecoregions
and within each major temperate biome (table 1).
On the other hand, three of the five taxa showed
much weaker or no correlations, and the scatter
in all of these relationships (and thus their
unpredictability) was high for all taxa.
On the more practical level of choosing areas
for conservation investment, the results are even
less encouraging. Because of the scatter mentioned
above, the statistical correlations found, even
when strongly significant, do not translate into
high overlap between priority sets based on
richness and endemism (table 3, fig. 2). A good
example is the relationship between the overall
richness and endemism indices (fig. 2F); the
statistical correlation between them is quite high
(table 1), but their priority sets overlap in only
2 out of a possible 11 ecoregions (table 3). This
contradiction is best understood by examining
figure 2F; although the two variables are
correlated overall, only two ecoregions fall in the
top decile for both richness and endemism.
Indeed, for all taxa except reptiles this overlap is
less than 50% (table 3). Basing conservation
strategies on richness, therefore, will seldom
effectively conserve areas of high endemism.
Previous studies on this topic also show a
mixture of results. In North America, KERR (1997)
found relatively strong correlations between
richness and endemism in four taxa: mammals, a
bee genus, a moth subfamily, and a butterfly
family. PRENDERGAST (1993), however, reported little
concordance between species–rich hotspots and
Ricketts
rare species in Great Britain, using birds,
butterflies, dragonflies, liverworts, and aquatic
angiosperms. Similarly, CEBALLOS et al. (1998) found
“very low correspondence” among areas of high
mammalian richness and endemism in Mexico.
What accounts for the differences in results
among these studies? Among other factors,
results may be influenced by the taxa and region
considered, the scale of observation (both extent
and resolution, LEVIN, 1992; PRENDERGAST et al.,
1993), the definition of endemism used, and the
choice of geographic units. For example,
PRENDERGAST (1993) based their analyses on 10 km
grid squares in Great Britain, while KERR (1997)
used much larger (2.5º of latitude and longitude)
grids over a much larger extent in North America
(in addition to testing different taxa). Clearly
the four studies (i.e., the three mentioned above
and mine) differ among themselves in several of
these factors, making it difficult to glean general
lessons from the collective results.
Perhaps of most interest are the contrasting
findings between my study and that of KERR
(1997). These two studies were performed in the
same region at similar scales, with one taxon in
common (mammals). Nevertheless, KERR (1997)
found high correlation in mammals (r = 0.807,
p < 0.001), while the results presented here show
quite a weak relationship (table 1). This difference
may be due to differences in the definition of
endemism. KERR (1997) calculates the endemism
value of a given square by summing, over all
species present in the square, the inverses of the
number of squares occupied by each species
(e.g., 1/24+1/137+1/3…). This measure, however,
is not independent of richness; the more species
present, the more inverses are added to the sum.
In contrast, counting the simple number of true
endemics in an area (i.e., species found nowhere
else) is not statistically related to richness
measures, and thus may better reveal the actual
relationship between these two measures of
conservation priority.
One caveat deserves mention here. Since a
typical species range overlaps with several
ecoregions (and thus ecoregions do not accrue
their richnesses independently), these richness
data probably contain a certain degree of spatial
autocorrelation (JONGMAN et al., 1995). This
problem tends to inflate the degrees of freedom
used in significance testing, and therefore the
probabilities reported here should be interpreted
with caution. However, these results remain
useful for comparing strengths of relationships
among taxa, because the correlation coefficients
themselves are unaffected (only the significance
tests). In addition, endemism, by definition, does
not suffer this same problem.
In conclusion, the results presented here and
in other studies (PRENDERGAST et al., 1993; KERR,
1997; C EBALLOS et al., 1998) indicate that
conservation biologists may not have the luxury
of assuming that management plans based on
Animal Biodiversity and Conservation 24.1 (2001)
“hotspots” of species richness will also capture
important centers of endemism. Additional
studies undertaken at different scales and with
different taxa may yield a better understanding
of the factors that determine the degree of
concordance between richness and endemism
patterns. Until then, however, conservation
biologists and land managers will need to
continue to balance these dual, and often
conflicting, goals of biodiversity conservation.
Acknowledgements
I thank the Conservation Science Program at World
Wildlife Fund–U.S. for their collaboration in originally
compiling these data. K. Bowen, J. Fay, M. Mayfield,
and J. Schwan helped to improve, error check, and
manage the databases. C. Boggs, K. Carney, J.
Hellmann and J. Hughes provided valuable
discussions and comments on the manuscript. Finally,
the support of U.S. NASA and The Summit
Foundation are gratefully acknowledged.
References
BIBBY, C. J., 1992. Putting biodiversity on the
map: priority areas for global conservation.
ICBP, Washington, DC.
CARROLL, S. S. & PEARSON, D. L., 1998. Spatial
modeling of butterfly species richness using
tiger beetles (Cicindelidae) as a bioindicator
taxon. Ecological Applications, 8: 531–543.
CEBALLOS, G., RODRIGUEZ, P. & MEDELLIN, R. A.,
1998. Assessing conservation priorities in
megadiverse Mexico: Mammalian diversity,
endemicity, and endangerment. Ecological
Applications, 8: 8–17.
DAILY, G. C. & EHRLICH, P. R., 1996. Nocturnality
and species survival. Proceedings of the
National Academy of Sciences of the United
States of America, 93: 11,709–11,712.
EHRLICH, P. R. & WILSON, E. O., 1991. Biodiversity
studies: science and policy. Science, 253: 750–752.
ESWG, 1995. A national ecological framework for
Canada. Agriculture and Agri–food Canada,
Research Branch, Centre for Land and Biological
Resources Research; and Environment Canada,
State of the Environment Directorate, Ecozone
Analysis Branch, Ottawa/Hull.
GALLANT, A. L., BINNIAN, E. F., OMERNIK, J. M. &
SHASBY, M. B., 1995. Ecoregions of Alaska. U.S.
Geological Survey, Washington, DC.
JONGMAN, R. H. G., TER BRAAK, C. J. F. & VAN
TONGEREN, O. F. R. (Eds.), 1995. Data analysis in
community and landscape ecology. Cambridge
University Press, New York.
KERR, J. T., 1997. Species richness, endemism, and
the choice of areas for conservation.
99
Conservation Biology, 11: 1,094–1,100.
LEVIN, S. A., 1992. The problem of pattern and
scale in ecology. Ecology, 73: 1,943–1,967.
MYERS, N., MITTERMEIER, R. A., MITTERMEIER, C. G.,
FONSECA, G. A. B. DA & KENT, J., 2000. Biodiversity
hotspots for conservation priorities. Nature,
403: 853–858.
OLSON, D. M. & DINERSTEIN, E., 1998. The Global
200: A representation approach to conserving
the Earth’s most biologically valuable
ecoregions. Conservation Biology, 12: 502–515.
OMERNIK, J. M., 1995. Level III ecoregions of the
continental US. US Environmental Protection
Agency, Washington, DC.
PEARSON, D. L. & CASSOLA, F., 1992. World–wide species
richness patterns of tiger beetles (Coleoptera:
Cicindelidae): Indicator taxon for biodiversity and
conservation studies. Conservation Biology, 6:
376–391.
PRENDERGAST, J. R., QUINN, R. M., LAWTON, J. H.,
EVERSHAM, B. C. & GIBBONS, D. W., 1993. Rare
species, the coincidence of diversity hotspots and
conservation strategies. Nature, 365: 335–337.
RICKETTS, T. H., DAILY, G. C. & EHRLICH, P. R. (in
press). Does butterfly diversity predict moth
diversity? Testing a popular indicator taxon at
local scales. Biological Conservation.
RICKETTS, T. H., DINERSTEIN, E., OLSON, D. M. &
L OUCKS, C., 1999a. Who’s where in North
America: patterns of species richness and the
utility of indicator taxa for conservation.
Bioscience, 49: 369–381.
RICKETTS, T. H., DINERSTEIN, E., OLSON, D. M., LOUCKS,
C., EICHBAUM, W., KAVANAGH, K., HEDAO, P.,
HURLEY, P. , CARNEY, K. M., ABEL, R. & WALTERS,
S., 1999b. Terrestrial ecoregions of North
America: A conservation assessment. Island
Press, Washington, DC.
ROSENZWEIG, M. L., 1995. Species diversity in space
and time . Cambridge University Press,
Cambridge.
SCOTT, M. J., DAVIS, F., CSUTI, B., NOSS, R., BUTTERFIELD,
B., GROVES, C., ANDERSON, H., CAICCO, S., D’ERCHIA,
F., EDWARDS, T. C. J., ULLIMAN, J. & WRIGHT, R. G.,
1993. Gap analysis: protecting biodiversity
using geographic information systems. Wildlife
Monographs, 123: 1–41.
SISK, T. D., LAUNER, A. E., SWITKY, K. R. & EHRLICH, P.
R., 1994. Identifying extinction threats: Global
analyses of the distribution of biodiversity
and the expansion of the human enterprise.
BioScience, 44: 592–604.
WEAVER, J. C., 1995. Indicator species and scale of
observation. Conservation Biology, 9: 939–942.
WILLIAMS, P. H. & GASTON, K. J., 1994. Measuring
more of biodiversity: Can higher–taxon richness
predict wholesale species richness? Biological
Conservation, 67: 211–217.
ZAR, J. H., 1999. Biostatistical analysis. 4th Edition.
Prentice Hall, Upper Saddle River, NJ.